Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{177885,
        author = {Harish K and S Balaji and K H Gayathri and Abishek V P and Asvindhan E and Madeshwaran R},
        title = {A Hybrid AI Framework: CNN for Malaria Detection and an Ensemble of Four Classifiers for Heart Disease Prediction},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {1992-1999},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=177885},
        abstract = {This research investigates the integration of artificial intelligence (AI) in the field of medical diagnostics by developing two intelligent models aimed at early disease detection and severity assessment. The first model is based on a Convolutional Neural Network (CNN), designed to analyse microscopic blood smear images and differentiate between malaria-infected and healthy cells. With an achieved accuracy of 93%, the model demonstrates high potential in assisting rapid, accurate malaria diagnosis. Beyond malaria, the architecture is adaptable and can be fine-tuned for binary classification tasks across other imaging-based diagnostic scenarios. The second model employs an ensemble learning strategy that combines the predictive strengths of Random Forest, XGBoost, and LightGBM classifiers. Trained on structured clinical datasets containing patient vitals and diagnostic parameters, this model effectively predicts the severity of heart disease by classifying patients into five distinct categories: No Disease, Mild, Moderate, Severe, and Critical. It achieves an accuracy of 85%, showcasing robust performance in handling complex, multidimensional tabular data. Together, these models lay the groundwork for a scalable and modular AI-powered diagnostic system capable of processing both visual and numerical medical inputs. When integrated into medical laboratory information systems, these models can enable real-time, automated interpretation of test results, thus reducing diagnostic turnaround times, minimizing human error, and enhancing the overall quality of care. Evaluation using key performance indicators such as precision, recall, F1-score, and ROC-AUC demonstrates the models’ reliability and competitiveness against traditional diagnostic approaches. The work underscores the transformative potential of AI in augmenting clinical workflows, enabling earlier interventions, and improving healthcare accessibility.},
        keywords = {},
        month = {May},
        }